npario/Qwen3.8-27B-OBLITERATED

VISIONPricing:Input $1.6 / Cached $0.15 / Output $12Concurrent Unit Cost:2Model Size:27BQuant:FP8Context Size:32kTool Calling:SupportedPublished:Sep 4, 2026License:apache-2.0Architecture:Transformer Open Weights Featherless Exclusive Cold

The npario/Qwen3.8-27B-OBLITERATED model is a 27 billion parameter language model based on the Qwen3.8 architecture, developed by Pliny the Prompter. This model is specifically engineered for genuine liberation from safety guardrails, providing direct answers without refusals or soft deflections. It maintains near-stock capability with a modest 2.1 percentage point drop in MMLU, excelling in code generation and restricted queries. Its primary use case is for research into refusal geometry, red-teaming, and AI safety evaluation requiring an unrestricted baseline.

Loading preview...

Qwen3.8-27B-OBLITERATED: Uncensored Language Model

This model, developed by Pliny the Prompter, is a 27 billion parameter variant of the Qwen3.8 architecture, distinguished by its surgical removal of safety guardrails. It aims to provide genuinely uncensored responses, eliminating both hard refusals and soft deflections (safety lectures) that are common in stock models. This "OBLITERATED" version (V3) achieves this through an advanced iterative refinement and targeted surgery process, building on previous versions' techniques like complementary abliteration blending.

Key Capabilities & Differentiators

  • Genuine Liberation: Provides direct answers to restricted queries, offering real substance over safety lectures.
  • High Code Generation Performance: Achieves 20/20 on cyber/code tasks, delivering functional implementations.
  • Near-Stock Capability: Maintains strong performance with only a -2.1 percentage point MMLU score reduction compared to the base Qwen3.8-27B.
  • Optimized Settings: Recommends specific inference parameters (temperature 0, repetition_penalty 1.15, empty system prompt) for optimal uncensored output.
  • Advanced Abliteration: Utilizes a sophisticated multi-stage weight-space surgery process to remove refusal behaviors while preserving core capabilities.

Ideal Use Cases

  • Alignment Research: For studying refusal geometry and safety robustness in LLMs.
  • Red-Teaming: Evaluating post-training safety mechanisms against weight surgery.
  • AI Safety Evaluation: Providing an unrestricted baseline for comprehensive assessments.
  • Local-First Users: For those requiring full control and unconstrained output on their own hardware.

Finetunes
2 models